Article: AMD's Mobile Strategy
By: Paul A. Clayton (paaronclayton.delete@this.gmail.com), December 15, 2011 8:20 pm
Room: Moderated Discussions
David Kanter (dkanter@realworldtech.com) on 12/15/11 wrote:
---------------------------
[snip]
>It's also worth pointing out that x86 decode is only a real
>pain when you talk about wide decode. A scalar x86 isn't all
>that hard, since you always know where to look for your
>instruction : )
I did not think scalar x86 was that decode friendly, either.
The register names are not in a consistent location and can
even be split between a prefix and the "ordinary" placement.
Finding the start of instructions can be substantially
simplified with predecode adding a continue/start bit for
each byte. (Intel seems to manage even without this
assistance.)
---------------------------
[snip]
>It's also worth pointing out that x86 decode is only a real
>pain when you talk about wide decode. A scalar x86 isn't all
>that hard, since you always know where to look for your
>instruction : )
I did not think scalar x86 was that decode friendly, either.
The register names are not in a consistent location and can
even be split between a prefix and the "ordinary" placement.
Finding the start of instructions can be substantially
simplified with predecode adding a continue/start bit for
each byte. (Intel seems to manage even without this
assistance.)